JP7832096B2 - Safety Assurance Support System and Safety Assurance Support Method - Google Patents

Description

This invention relates to a safety assurance support system and a safety assurance support method.

Regarding technology for verifying the use of safety harness hooks during work at height, there is a technology disclosed in Patent Document 1 below. Patent Document 1 states that "the management system comprises: an image acquisition unit that acquires on-site images of a work site where a worker equipped with a safety harness is working; an image processing unit that generates color images of the color attached to the safety harness hook and the color attached to the handrail or lifeline at the site from the on-site images; and a determination unit that determines the usage status of the hook based on the color images of the hook and the handrail or lifeline."

Japanese Patent Publication No. 2021-129953

Here, some work at heights, such as maintenance work on top of an elevator car, is performed in a dark environment using lighting. In this case, with the configuration described in Patent Document 1, the color information of the hook and locking part (handrail or lifeline) cannot be recognized as intended due to the influence of lighting, making it easy for misjudgments to occur due to missed detections.

Therefore, the present invention aims to provide a safety support system and a safety support method that can prevent misjudgment of the use of a safety harness hook, even in dark environments.

To solve the above problems, for example, the configuration described in the claims may be adopted.
The present invention includes several means for solving the above problems, but to give one example, it includes a management device that detects the hook of a safety harness worn by a worker in images sequentially transmitted from a camera worn by the worker, and instructs an output terminal device carried by the worker to output an alert if the hook is not detected, the hook is provided with a retroreflective material having a specific hue on its surface, and the management device is a safety assurance support system that detects the worker's movement to enter the work site where the hook is used based on images sequentially transmitted from the camera, and detects the hook based on the color information of each image after the entry movement is detected.

This invention provides a safety support system and method that can prevent misjudgments of hook usage by improving the detection accuracy of safety harness hooks, even in dark environments.

This is a device configuration diagram of the safety support system according to the embodiment. This figure shows an example of the installation of the safety support system according to the embodiment. This is a diagram illustrating the upper part of the elevator car, which will serve as the work site. This figure shows an example of a warning sign installed at a work site. This is a functional configuration diagram of the control device. This is a flowchart showing the safety support method according to the embodiment. This figure shows an example of a worker's actions at a work site. This diagram shows the change in hue of a captured image due to the illumination of light.

The following describes in detail embodiments of the safety support system and safety support method of the present invention with reference to the drawings. The safety support system and safety support method described herein are systems and methods for supporting the safety of workers during work at height in dark environments. In the following embodiments, work on top of an elevator car is used as an example of work at height in a dark environment, but the invention is not limited to this.

≪Safety Assurance Support System≫
Figure 1 is a device configuration diagram of the safety support system 1 according to an embodiment. Figure 2 is a diagram showing an example of the safety support system 1 being installed according to an embodiment. First, before explaining the device configuration of the safety support system 1, we will explain the equipment of the worker [W] engaged in work at height in a dark environment and the work environment.

<Equipment of worker [W]>
As shown in Figure 2, workers [W] engaged in work at heights are required to wear a full-body harness type safety belt 31 and a helmet 32. In particular, if the work environment is dark, workers [W] will hold a light 33 attached to the helmet 32.

Here, the safety harness 31 is a fall prevention device comprising a belt mechanism 31a, a rope 31b, and a hook 31c. The belt mechanism 31a is a single integrated structure consisting of shoulder belts, a chest belt, a waist belt, thigh belts, and a pelvic belt. The rope 31b has a buffering mechanism and its base end is attached to the back side of the shoulder belts in the belt mechanism 31a. As shown in the enlarged view in Figure 2, the hook 31c is a locking device attached to the end of the rope 31b. This hook 31c is used by locking it to a locking part provided at the work site, and in other cases by locking it to the chest belt of the belt mechanism 31a.

In particular, this hook 31c has a surface covered with retroreflective material. Normally, hooks 31c are often made of semi-glossy steel and are susceptible to ambient light, meaning their appearance can vary significantly depending on the type of light source and the angle of incidence. Therefore, the hook 31c used in this safety support system 1 has a surface covered with retroreflective material. Retroreflective material is a material that reflects light from a light source back towards the light source, allowing it to reflect light from the light 33 back towards the light 33 even in dark environments. Such a hook 31c may be made by wrapping retroreflective tape around the original hook 31c, as shown in the enlarged view in Figure 2, and it is preferable that as much of the surface area as possible is covered with retroreflective material.

Furthermore, the color of the retroreflective material on the surface of the hook 31c is a specific hue not present in the work area, thereby preventing false detection of the hook 31c. Therefore, if the work area is on top of an elevator car, a hue not present on the car is applied as the color of the retroreflective material. Examples of suitable hues for such retroreflective materials include red, blue, and green. In particular, red may react to human faces or incandescent light, and blue may react to protective sheets (blue tarpaulins), therefore, green is preferred.

<Work Environment>
Figure 3 is a diagram illustrating the upper part of an elevator car, which serves as the work site at height. As shown in Figure 3, the elevator car 41, which serves as the work site, is suspended by a main rope 43 within a hoistway 42 that extends vertically along the building. The elevator car 41 is equipped with a car door 41a for boarding. The worker [W] (see Figure 2) opens a landing door (not shown) located in the hoistway 42, enters the hoistway 42 through this landing door, and boards the upper part 41b of the elevator car 41, which serves as the work site. The hoistway 42 is a dark environment with little to no lighting.

Furthermore, the work area on the cage 41b is provided with a locking mechanism 401 for securing the hook 31c of the safety harness 31. The locking mechanism 401 includes, for example, a handrail 401a erected on the cage 41b, an eyebolt 401b fixed to the floor surface of the cage 41b, and a U-shaped hook (not shown in this illustration). In addition, the cage 41b has a fall prevention fence (not shown in this illustration) around its perimeter, and various pieces of equipment are fixed to the floor surface.

Furthermore, a warning sign 402 is provided on the top of the elevator car 41b to alert the worker [W]. The warning sign 402 is intended to remind the worker [W], who is boarding the elevator car 41b, of the precautions to take while at a height. Such a warning sign 402 is installed on the floor of the elevator car 41b, specifically on the side where the worker [W] boards, i.e., on the elevator door (not shown) in the elevator shaft 42, on the elevator car door 41a side, so that it is always in the worker's [W] line of sight when they board the elevator car 41b. Figure 4 shows an example of a warning sign 402 installed at a work site. As shown in Figure 4, the warning sign 402 displays a warning text 402b and a warning mark 402c on a white or yellow background 402a to improve visibility in dark places.

Because this warning indicator 402 is located in a conspicuous place when the worker [W] is about to board the cage 41b and is facing the cage 41b, it is used as a "feature marker" for color information correction, as explained later in the safety support method.

<System Configuration>
Next, the device configuration of the safety support system 1 for workers [W] engaged in work at height in a dark environment will be described based on Figures 1 and 2. The safety support system 1 shown in Figures 1 and 2 comprises a camera 11, an output terminal device 12, and a management device 13. These are connected via a network 2 and are configured as follows.

[Camera 11]
Camera 11 is used to continuously capture images of the work site as video, and is attached to each worker [W]. Camera 11 also has a communication function and is configured to sequentially transmit the captured images, along with the identification information of each camera 11, to the management device 13 via the network 2.

The mounting position of the camera 11 relative to the worker [W] is not particularly limited, but it should be mounted in a way that allows it to capture images in the direction of the worker's [W] line of sight. From this perspective, it is preferable that the camera 11 be attached to the chest belt of the belt mechanism 31a or to the front side of the helmet 32, and preferably in a position close to the light 33. Therefore, it is preferable that the belt mechanism 31a or helmet 32 be provided with a mounting section for attaching the camera 11. Note that the light 33 attached to the helmet 32 may be integrated with the camera 11.

[Output terminal device 12]
The output terminal device 12 is a device for outputting alerts to workers [W], and it has identification information that is associated with the camera 11, and is used by being attached to each worker [W]. The output terminal device 12 also has a communication function that allows individual communication with the identification information, and is configured to output alerts based on instructions transmitted from the management device 13 via the network 2.

The mounting position of the output terminal device 12 on the worker [W] is not particularly limited, but the worker [W] will wear the camera 11 and the output terminal device 12, which are associated with each other by identification information. The output terminal device 12 may also have some of the functions of the management device 13 described below, and a camera function that captures images of the work site and transmits them to the output terminal device 12. If the output terminal device 12 has a camera function and also functions as the camera 11, it is preferable that the output terminal device 12 be mounted in a way that allows it to capture images in the direction of the worker [W]'s line of sight, for example, by attaching it to the chest belt of the belt mechanism 31a. In this case, it is preferable that the belt mechanism 31a is provided with a mounting section for attaching the output terminal device 12, which also functions as the camera 11.

[Management device 13]
The management device 13 is a terminal device having a computer for implementing safety assurance support methods and is located, for example, in a management center for elevator maintenance and management. In addition to the computer, the management device 13 may also have a display unit and an operation unit, and may be a so-called personal computer. The computer in the management device 13 is equipped with a CPU (Central Processing Unit), RAM (Random Access Memory), non-volatile storage units such as ROM (Read Only Memory), and a network interface. Such a management device 13 continuously and sequentially receives images captured by the camera 11 via the network 2. It also determines whether or not to output an alert based on the received video and sends an instruction to output an alert to the output terminal device 12 via the network 2.

Figure 5 is a functional configuration diagram of the management device 13. As shown in Figure 5, the management device 13 includes, as functional units for implementing the safety assurance support method, an image holding unit 131, a blackout detection unit 132, an elapsed time count unit 133, a feature marker detection unit 134, a hook detection unit 135, a color information correction unit 136, a hook re-detection unit 137, and an alert output instruction unit 138. The functions performed by each functional unit will be described below with reference to Figures 1 to 4, along with Figure 5.

-Image holding unit 131-
The image holding unit 131 stores the images transmitted sequentially from the camera 11 as a continuous video. Each image is associated with the identification information of the receiving camera 11. Of the stored images, the image holding unit 131 stores the images in which a hook has been detected by the hook detection unit 135 or the hook re-detection unit 137, which will be described later, as hook detection images.

-Blackout detection unit 132-
The blackout detection unit 132 determines whether the overall brightness of the image transmitted from the camera 11 has exceeded a predetermined threshold, that is, whether the image has gone black. Details of the determination made by the blackout detection unit 132 will be explained in the safety support method described below.

- Elapsed time counting unit 133 -
The elapsed time counting unit 133 counts the elapsed time since the blackout determination unit 132 determined that the image had gone black.

-Feature marker detection unit 134-
The feature marker detection unit 134 sequentially analyzes the images transmitted sequentially from the camera 11 and detects feature markers within each image. Here, a feature marker is a display used as a reference for correction processing by the color information correction unit 136, which will be described next. Such feature markers are standardly installed at the work site, are always visible to workers [W] when they enter the work site, and have good visibility in dark places. As an example of such a feature marker, a warning display 402 (see Figures 3 and 4) can be used.

- Hook detection unit 135 -
The hook detection unit 135 sequentially analyzes the images transmitted from the camera 11, detects the hook 31c in each image, and determines whether or not the hook 31c has been detected. Details of the determination made by the hook detection unit 135 will be explained in the safety support method described below.

-Color information correction unit 136-
The color information correction unit 136 corrects the color information used as the basis for the hook detection unit 135's judgment when the hook detection unit 135 determines that the hook 31c has not been photographed. Details of the color information correction performed by the color information correction unit 136 will be explained in the safety assurance support method below.

- Hook re-detection unit 137 -
The hook re-detection unit 137 performs hook re-detection for each image held in the image holding unit 131, which is an image in which hook detection was performed by the hook detection unit 135, based on the color information corrected by the color information correction unit 136. The hook re-detection unit 137 performs re-detection of the hook 31c within each image by analyzing each image based on the corrected color information, and determines whether or not the hook 31c has been detected. Details of the determination made by the hook re-detection unit 137 will be explained in the safety assurance support method below.

- Alert output instruction unit 138 -
The alert output instruction unit 138 sends an alert output instruction indicating that the hook was not used to the output terminal device 12 associated with the camera 11 if the hook re-detection unit 137 does not detect a hook 31c in any of the images it has detected. The alert output instruction unit 138 maintains the identification information of the camera 11 and the identification information of the output terminal device 12 in association with each other.

≪Safety ensuring support method≫
Next, an embodiment of the safety support method implemented by the safety support system 1 described above will be explained.

Figure 6 is a flowchart illustrating a safety support method according to an embodiment. The procedure shown in this flowchart is a procedure to support the safety of a worker [W] during high-altitude work in a dark environment by checking the usage status of the safety harness hook, according to the program of the management device 13 in the safety support system 1 described using Figure 1. When entering the work site, the worker [W] wears a safety harness 31, helmet 32, light 33, camera 11, and output terminal device 12, as shown in Figure 2. At this time, the worker [W] secures the hook 31c of the safety harness 31 to the chest belt of the belt mechanism 31a of the safety harness 31. Then, the worker [W] turns on the light 33, camera 11, and output terminal device 12, boards the elevator car which is the work site, and begins work while self-securing using the hook 31c on top of the car.

The following describes the procedure for supporting safety during work at height in a dark environment, following the flowchart in Figure 6 and referring to Figures 1 to 5. Note that the flowchart in Figure 6 begins when the management device 13 receives an image transmitted from the camera 11 worn by the worker [W].

<Step S101>
In step S101, the image holding unit 131 starts recording the images transmitted sequentially from the camera 11. The image holding unit 131 saves the images transmitted sequentially from the camera 11 as a continuous video. Each image is associated with the identification information of the receiving camera 11.

<Step S102>
In step S102, the blackout determination unit 132 determines whether the brightness of the entire image transmitted from the camera 11 has become darker than a predetermined determination value, that is, whether the image has blacked out. In this case, the blackout determination unit 132 uses the brightness of the entire image initially transmitted from the camera 11 as a reference value, and determines that the image has blacked out (YES) if the brightness of the entire image relative to the reference value has become darker than a predetermined difference value.

Here, the elevator car top 41b, which serves as the work area, is a dark environment within the hoistway 42. Therefore, when the worker [W], wearing the light 33, enters the elevator car top 41b from the landing, which has a certain level of brightness, facing into the hoistway 42, the amount of light received by the camera 11 decreases drastically. Consequently, when the difference in brightness of the images sequentially transmitted from the camera 11 exceeds a predetermined difference value and the image becomes dark, it can be determined that the worker [W] wearing the camera 11 has performed the action of entering the elevator car top 41b, which is the work area.

Furthermore, the brightness difference value used to determine when the screen has gone dark should be predetermined, for example, through preliminary experiments.

<Step S103>
In step S103, the elapsed time counting unit 133 starts counting the elapsed time from the point in step S102 when it was determined that the image had gone dark. This elapsed time is counted as the time elapsed from the point in time when the worker [W] started to get onto the basket 41b, that is, from the point in time when the worker started to enter the work site.

<Step S104>
In step S104, the feature marker detection unit 134 analyzes the image transmitted from the camera 11 and detects feature markers in the image captured by the camera 11. At this time, the feature marker detection unit 134 uses the warning display 402 (see Figures 3 and 4) as the feature marker. The feature marker detection unit 134 recognizes the string of the warning text 402b and the warning mark 402c of the warning display 402, which are stored in advance, from the image and detects them as feature markers, and stores the image containing the detected feature markers.

The warning display 402, used as a feature marker, is captured by the camera 11 when the worker [W] turns towards the cage 41b in preparation for boarding. Therefore, the feature marker is detected as part of the boarding scene onto the cage 41b. Consequently, the elapsed time counting unit 133 may begin counting the elapsed time based on the detection of the feature marker, as it indicates the worker [W] has entered the work area. In this case, even if the light intensity of the worker [W]'s light 33 is sufficiently bright, the system can correctly detect the worker's entry into the work area in a dark environment. Furthermore, the elapsed time counting unit 133 may also begin counting the elapsed time from the point when the image darkens and the feature marker is detected.

<Step S105>
In step S105, the hook detection unit 135 determines whether or not a hook 31c has been detected in the image based on the color information in the image transmitted from the camera 11. In this case, the hook detection unit 135 uses, for example, the HSV color model as the color information. The HSV color model, also known as the HSB model (B stands for Brightness), is a type of method for representing colors handled on a computer numerically, and expresses color using three elements: "hue,""saturation," and "value."

The hook detection unit 135 extracts pixels from the image that satisfy the predetermined judgment thresholds (upper and lower thresholds) for hue, saturation, and brightness. The hook detection unit 135 determines that a hook 31c has been detected (YES) if the number of extracted pixels exceeds a predetermined number of pixels. These judgment thresholds are set to include the hue, saturation, and brightness of the retroreflective material on the surface of the hook 31c when photographed, for example, under natural light. Furthermore, the number of pixels used as the judgment criterion for detecting the hook 31c is set based on the number of pixels of the hook 31c captured by the camera 11 when the worker [W] uses the hook 31c, according to the camera's field of view and pixel count.

Figure 7 shows an example of the actions of worker [W] at a work site. As shown in Figure 7, worker [W], having boarded the elevator car 41b which is the work site, detaches the hook 31c from the belt mechanism 31a of the safety harness 31 and locks the hook 31c to the locking part 401 on the elevator car 41b (see Figure 3). Because the hook 31c is detached from the belt mechanism 31a of the safety harness 31, it becomes easier for it to enter the field of view of the camera 11. Furthermore, in the action of locking the hook 31c to the locking part 401, worker [W] visually checks their hand holding the hook 31c, causing the camera 11 to move towards the hand holding the hook 31c, thus bringing the hook 31c into the camera 11's field of view. Therefore, when the hook 31c is detected in the image transmitted from the camera 11, it is considered that worker [W] has performed the action of locking the hook 31c to the locking part 401.

Therefore, if the hook detection unit 135 determines that it has detected the hook 31c (YES) according to the procedure described above, it determines that the worker [W] has performed the action of locking the hook 31c to the locking part 401 and proceeds to step S106. On the other hand, if the hook detection unit 135 determines that the hook 31c is not detected (NO), it determines that the worker [W] has not performed the action of locking the hook 31c to the locking part 401 and proceeds to step S107.

<Step S106>
In step S106, the image holding unit 131 records the image in which the hook 31c was detected in step S105 as the hook detection image and terminates the process.

<Step S107>
On the other hand, in step S107, the hook detection unit 135 determines whether the elapsed time counted by the elapsed time count unit 133 has exceeded a predetermined time. This predetermined time is within a range that sufficiently includes the time required from when the worker [W] gets onto the cage 41b, which is the work site, until the hook 31c is attached to the locking part 401 on the cage 41b, and is preferably as short as possible from the viewpoint of fall prevention. For example, let's say this predetermined time is set to 2 minutes.

The hook detection unit 135 proceeds to the next step S108 if it determines that a predetermined time has elapsed (YES). On the other hand, if it determines that the predetermined time has not elapsed (NO), it returns to step S105 and repeats the detection of the hook 31c for each image sequentially transmitted from the camera 11.

<Step S108>
In step S108, the color information correction unit 136 performs a process to correct the color information that serves as the criteria for hook detection by the hook detection unit 135. The color information corrected here refers to the judgment thresholds (upper and lower thresholds) of each color information that serves as the criteria for hook detection. In other words, in a dark environment, the image captured by the camera 11 is affected by the light emitted from the light 33, so correction is performed to eliminate this effect.

Figure 8 shows the change in hue of a captured image due to illumination, and is a figure of an image taken with a camera of a hook 31c having a retroreflective material on its surface. As shown in Figure 8, when the retroreflective material on the surface of the hook 31c is illuminated by natural light 1001 and incandescent light 1002, the hue of the hook 31c in the resulting image [P1] is almost natural and is the original hue of the retroreflective material. In contrast, when the retroreflective material on the surface of the hook 31c is illuminated particularly by LED 1003, the hue of the hook 31c in the resulting image [P2] may be a hue that reflects a stronger blue tint than the original hue of the retroreflective material. This is thought to be because some white LEDs have a blue LED inside as a light emission method, and due to manufacturing process constraints, etc., this blue light does not mix properly with the yellow phosphor in a balanced manner, resulting in the emission of blue light. Alternatively, it is also possible that the light's lens is causing spectral dispersion. In any case, when illuminated by a white LED light, the hue of the object may change (especially shifting towards blue). Therefore, in step S105, it is possible that the hook 31c could not be correctly detected due to the influence of the light emitted from the light 33.

Therefore, the color information correction unit 136 corrects the judgment thresholds (upper and lower thresholds) for each color information (hue, saturation, and brightness) that are pre-set for detecting the hook 31c, based on the image of the feature marker recognized in step S104. Specifically, first, it calculates the amount of change in the background color information of the feature marker recognized in step S104, relative to a reference value of the background color information of the feature marker. For example, hue is used as the color information. The reference value is, for example, the hue of the background of the feature marker when it is photographed by the camera 11 under natural light conditions. The color information correction unit 136 corrects the judgment thresholds (upper and lower thresholds) for hue among hue, saturation, and brightness in accordance with the calculated amount of change in hue. In other words, if the hue of the background of the feature marker has changed towards blue, the unit corrects the judgment thresholds (upper and lower thresholds) for hue by the amount of that change towards blue.

<Step S109>
In step S109, the hook re-detection unit 137 analyzes each image held in the image holding unit 131 based on the judgment threshold of the color information corrected by the color information correction unit 136, and makes a determination as to whether or not a hook 31c has been detected. At this time, the hook re-detection unit 137 analyzes each image held in the image holding unit 131 for each pixel from the time counting started in step S103 until it is determined in step S107 that a predetermined time has elapsed (YES). These images are the images for which the hook detection unit 135 has already detected a hook 31c in step S105. The procedure for analyzing each image is the same as the procedure in step S105.

The hook re-detection unit 137 detects an image in which the number of pixels satisfying the corrected judgment threshold exceeds a preset number of pixels. If it detects this, it determines that the hook 31c has been detected (YES), i.e., that the worker [W] used the hook 31c, and proceeds to step S106. As a result, the image holding unit 131 records the image in which the hook re-detection unit 137 determined that the hook 31c was detected as the hook detection image, and terminates the process.

On the other hand, if no image is found among the analyzed images that satisfies the corrected judgment threshold with a number of pixels greater than the preset number, the hook 31c is not detected (NO), meaning that the operator [W] is not using the hook 31c, and the process proceeds to step S110.

<Step S110>
In step S110, the alert output instruction unit 138 sends an alert output instruction for the unused hook to the output terminal device 12 associated with the camera 11 that transmitted the image, and terminates the process. As a result, the worker [W] wearing the output terminal device 12 is notified of the unused hook 31c alert. This alert is issued immediately after a predetermined time (2 minutes in this case) has elapsed since entering the work site, prompting the worker [W] to secure themselves by using the hook 31c.

<<Effects of the Embodiment>>
According to the embodiment described above, the system detects the entry of a worker [W] into the work site where the hook 31c is to be used, based on images transmitted sequentially from the camera, and then detects the hook 31c, which has a retroreflective material on its surface. This makes it possible to improve the detection accuracy of the hook 31c in dark work environments without providing light sensors or load sensors on the hook 31c or the locking part 401, and prevents misjudgment of the use of the hook 31c.

Furthermore, if the hook 31c is not detected within a predetermined time period after the detection of entry, the system corrects the detection threshold (color information) and re-detects the hook 31c. This prevents detection failures of the hook 31c due to the influence of light from the light 33, improving the detection accuracy of the hook 31c. It also prevents misjudgment of hook 31c usage regardless of the type of light 33 used in dark environments.

Furthermore, the correction of the detection threshold (color information) was performed when hook 31c could not be detected between the detection of entry and the elapsed time. This also has been shown to suppress the increase in over-detection of hook 31c when correction is performed immediately after the detection of the feature marker in step S104.

Furthermore, the system is configured to correct the judgment threshold (color information) based on the image of the warning display 402, which is always within the field of view of the worker's camera 11 when entering the work site. Therefore, there is no need to individually prepare color patterns for correction, and it is possible to improve hook detection accuracy and prevent misjudgments without burdening the worker [W] with photographing color patterns.

Furthermore, the present invention is not limited to the embodiments and modifications described above, but includes a wide variety of further modifications. For example, the embodiments described above are explained in detail for clarity and are not necessarily limited to those comprising all the described configurations. Also, it is possible to replace parts of the configuration of one embodiment with those of another embodiment, and it is also possible to add configurations from other embodiments to the configuration of one embodiment. Furthermore, it is possible to add, delete, or replace parts of the configuration of each embodiment with those of other embodiments.

1...Safety support system 11...Camera 12...Output terminal device 13...Management device 31...Safety harness 31c...Hook 41b...On top of cage (work site)
133... Elapsed time counting unit 135... Hook detection unit 134... Feature marker detection unit 136... Color information correction unit 137... Hook re-detection unit 138... Alert output instruction unit 402... Warning display (feature marker)
402a... Substrate [W]... Worker

Claims (9)

The system includes a management device that detects the hook of a safety harness worn by a worker in the images sequentially transmitted from a camera worn by the worker, and instructs an output terminal device carried by the worker to output an alert if the hook is not detected.
The aforementioned hook is provided with a retroreflective material having a specific hue on its surface.
The management device is a safety support system that detects the worker's entry into the work site where the hook is used based on images transmitted sequentially from the camera, and detects the hook based on the color information of each image after the entry has been detected. The security support system according to claim 1, wherein the management device detects the entry action when the brightness of the images transmitted sequentially from the camera darkens within a predetermined range. The safety support system according to claim 1, wherein the management device detects pixels having the hue of the retroreflective material in each image after detecting the entry action, and determines that a hook has been detected when the number of detected pixels exceeds a predetermined number. The aforementioned control device is
A time elapsed counting unit that counts the elapsed time since the detection of the entry action,
A hook detection unit performs hook detection on each image among the images sequentially transmitted from the camera, between the detection of the entry action and the time elapsed count value in the elapsed time count unit until a predetermined time has elapsed.
The safety support system according to claim 1, further comprising: an alert output instruction unit that instructs the output of the alert when the hook detection unit does not detect the hook. The aforementioned control device is
A feature marker detection unit detects characteristic feature markers of the work site from among the images sequentially transmitted from the camera,
If the hook is not detected by the hook detection unit, the color information correction unit corrects the judgment threshold of the color information used for detecting the hook based on the feature marker detected by the feature marker detection unit.
The safety support system according to claim 4, further comprising: a hook re-detection unit that performs hook re-detection for each image in which the hook detection unit has performed detection based on the color information corrected by the color information correction unit. The aforementioned characteristic marker is a warning sign installed at the work site,
The safety support system according to claim 5, wherein the color information correction unit corrects the hue judgment threshold of the color information based on the hue of the background of the warning display in the image transmitted from the camera. The safety support system according to claim 1, wherein the management device records an image of the detected hook. The safety support system according to claim 1, wherein the retroreflective material has a green hue as the specific hue on its surface. A safety support method that detects the hook of a safety harness worn by a worker in images sequentially transmitted from a camera worn by the worker, and alerts an output terminal device carried by the worker if the hook is not detected,
As the aforementioned hook, one is provided with a retroreflective material having a specific hue on its surface.
A safety support method for detecting the hook, which involves detecting the worker's entry into the work site where the hook is used based on images transmitted sequentially from the camera, and then detecting the hook based on the color information of each image after the entry has been detected.